At Arm's Length: Controlling State-Funded Militants in Civil War

Despite a $500 million investment, most Syrian militants receiving U.S. sponsorship did not attack ISIL. This failure is alarming; as civil wars begin to include more militant groups, states are increasingly interested in influencing their behavior in order to advance national interests. It is also surprising the degree to which militants are able to shirk given the power disparity between states and militants. This thesis asks why strong states struggle to control the weak militant groups they fund in Syria. Most answers rely on principal-agent theory alone, but I incorporate more recent literature on competition and alliances within insurgencies. I also employ a mixed-method analysis of 350 attacks conducted by Sunni militant groups with U.S. or Saudi sponsorship from 2014 to 2016.

Militant group adherence to state goals is highly mediated by participation in alliances. The extent to which Syrian militant groups in alliances obey their sponsors is influenced by the allocation of power within alliances and alliance members’ preferences. Curiously, militant groups are more inclined to follow state sponsor objectives when they ally with other militants who have different aims. Unlike alliances composed entirely of extremists, militant groups can credibly threaten to defect from alliances with groups that have dissimilar goals. In response, these alliances choose a narrow range of targets that conform to the preferences of all members and their state sponsors. This suggests that to avoid further failures, state sponsorship programs must place more consideration on alliances within insurgencies.